Quick answer

This index tracks 32GB (2x16GB) DDR5 kits at named direct US retailers, with Amazon excluded by design. In the day-one capture on 2026-08-12, the cheapest DDR5-6000 kit passing every verification rule was $449.99 at B&H — about $14.06 per GB, roughly 4x its 2024 street price. Open CC BY 4.0 data.

By LK Wood IV · Published 2026-08-12 · ~5 min read · St. Louis County, MO

The short version

DDR5 prices roughly tripled during the 2026 memory supply crisis, and the numbers that circulate are mostly screenshots without dates or methods. This page is the start of a dated, checkable record: a $/GB index for mainstream 32GB (2x16GB) DDR5 kits observed at named direct US retailers only, with the method, the acceptance rules, every rejection reason, and the raw data published alongside the numbers.

It is labeled a pilot because the index metric has an activation gate it has not passed yet — and rather than loosen the rules to manufacture a number, the gate status is printed here verbatim.

Day-one observations (2026-08-12)

KitCheapest qualifying offer$/GBRetailer
Corsair Vengeance 32GB (2x16) DDR5-6000 CL30$449.99$14.06B&H
Corsair Vengeance RGB 32GB (2x16) DDR5-6000 CL30$449.99$14.06B&H
Corsair Dominator Titanium 32GB (2x16) DDR5-6000$649.99$20.31Newegg

21 qualifying offers total across 3 kits, out of 546 listings reviewed. For scale: the Vengeance 6000 CL30 kit was a $95–110 street-price part through 2024 — today’s cheapest direct-retailer offer is roughly 4x that.

Gate status, verbatim: 3 of 12 tracked kits produced qualifying offers; the index metric (median cheapest $/GB across kits) activates at 6. Direct retailer stock is not independently confirmed at the listing page yet, so every row carries verification_status: discovery_only.

What gets rejected, and why that’s the point

Of 546 listings reviewed in the day-one capture, 525 were rejected, each with recorded reasons: marketplace and third-party sellers (the largest bucket), listings that don’t state the 32GB capacity or the 2x16GB split in the title, wrong speed for the tracked SKU, used/renewed condition, and prices outside a sanity band. A price index built on listings a script guessed about would be smoother — and worthless. The rejection log ships with the dataset.

Method, exactly

Captures run against Google Shopping (US, English) via DataForSEO’s merchant API, one exact-model query per tracked kit. An offer qualifies only if the seller matches a named direct-retailer allowlist (Best Buy, Newegg, B&H, Micro Center, Adorama, Walmart first-party, and peers — marketplace suffixes excluded), the title states the exact kit line, speed, 32GB capacity, and 2x16GB split (a “Dual Channel” declaration on a 32GB listing counts as the split), condition is new, and the USD price sits inside a $60–900 band. The band’s ceiling was raised from $500 on day one when real direct-retailer offers at $469–499 for the cheapest kit proved the old ceiling was written for a market that no longer exists — that correction is recorded in the change log, not hidden.

Amazon is excluded by design (see FAQ). The full collector, product basket, tests, raw responses, and rejection logs are in the site’s repository.

Use the data (CC BY 4.0)

Download the day-one observations: ram-price-observations.csv — one row per qualifying offer, with capture date, kit, price, $/GB, seller, and verification status. Once the activation gate opens, the daily index series publishes at /data/ram-price-index.csv (median cheapest $/GB across kits, one row per UTC day). License: Creative Commons Attribution 4.0 — attribute as “TechFuelHQ DDR5 RAM Price Index” linking to this page.

Context for the crisis itself: the RAM & SSD price crisis explainer and the DDR5 buying guide — the guide’s price-tracking data is what earned this site its first external news citation, and this index is that work made repeatable.

Change log

  • 0.9.0 (2026-08-12) — pilot capture and method publication. Four method corrections made against measured evidence on day one, all recorded: price ceiling $500 → $900 (285 real listings died on the stale band); “Dual Channel” accepted as an explicit module-split marker (Newegg writes Trident Z5 Neo kits this way); per-product offer de-duplication (a Corsair offer was being consumed and mis-rejected under another kit’s query); basket v1.1 swapped two kits with zero direct-retailer presence (ADATA XPG Lancer, Patriot Viper Venom) for Dominator Titanium 6000 and Flare X5 5600.

Frequently asked questions

What does 32GB of DDR5 cost right now?
In this dataset’s day-one capture (2026-08-12), the cheapest 32GB (2x16GB) DDR5-6000 kit that passed every rule — named direct US retailer, new condition, explicit kit configuration in the listing title — was $449.99 (about $14.06/GB) for Corsair Vengeance at B&H. Corsair’s Dominator Titanium 6000 kit was $649.99 at Newegg. Marketplace and third-party-seller listings, which are sometimes cheaper and often less reliable, are excluded by design.
Why is Amazon excluded from this index?
Two reasons. Amazon Associates policy only permits displaying Amazon price data through Amazon’s own live mechanisms, so a dated open snapshot of Amazon prices cannot be published — the same policy that led this site to remove prices from its PC parts dataset on 2026-08-01. And Amazon listings mix marketplace sellers with first-party stock in ways a mechanical rule cannot reliably separate. Direct retailers with named storefronts make each observation checkable.
Why does the index show only 3 of 12 kits today?
Because the acceptance rules are strict and public: the listing must come from a named direct retailer, state the 32GB capacity and the 2x16GB module split explicitly, name the exact kit line and speed, be new condition, and carry a plausible USD price. In the day-one capture only Corsair’s lines produced qualifying direct-retailer offers; G.Skill, Kingston, TeamGroup, and Crucial 2x16 kits surfaced mostly through marketplace sellers or ambiguous listings, which are rejected rather than guessed at. The index metric activates at 6 qualifying kits.
How often is this captured?
The collector is built for daily capture, one row per UTC day once the activation gate opens. Every capture stores its complete raw API response, every accepted offer, and every rejected offer with the exact reasons, so any published number can be traced to the listing it came from.

Evidence ledger

Last updated
Methodology
This guide was written and edited by Lowell K. Wood IV in St. Louis County, MO. Specs, prices, commands, and version numbers are drawn from the official vendor, reseller, and project documentation current on the date above, and were verified before publishing. First-person hardware claims appear only where the article shows a verifiable artifact — a photo, receipt, or measurement — or links to the TechFuelHQ Open Bench Datasets. Every fact is human-verified against its cited source before publishing; AI assists with first-draft structure and source-gathering, not with the verdict. Full editorial standard: methodology.
Update log
  • 2026-08-12 — Last reviewed and updated.
Corrections
Spotted an error or stale price? Email hello@techfuelhq.com. Confirmed corrections are added to the update log above.

About the author

Written by Lowell K. Wood IV. Lowell builds and runs TechFuelHQ from St. Louis, Missouri, pairing thirteen-plus years of hands-on homelab, PC, server, and networking experience with cited third-party testing and first-party benchmarks on the gear he still runs. He also works ground EMS as a Nationally Registered Paramedic (NREMT).